开云体育

In recent thread on IB board, Kent commented that C++ was the way to
go for creating serious apps. Is this really true any more if you're
starting from .NET, where all of its languages are based on same
underlying runtime lib.s and architecture. For purposes of building a
killer trading app, does it really matter much whether you're using
VB.NET or C#.NET or C++? And can't you just mix in legacy C and C++
lib.s as necessary?

I transfered this question to the VB message board I haunt.

Here is a response from one guy there:

Re: C++ vs. C#.NET vs. VB.NET
----------------------------------------------------------------------
----------
In .NET, pretty much all the languages are on equal footing. I think
VB is at a disadvantage with some of the OO stuff, though. I don't
think it has function overloading, which is extremely useful, and I'm
still not sure how much inheritance it allows. I could be wrong on
these, though. I haven't used VB.NET at all. 90% of code or more will
be almost exactly the same with VB and C# because of the shared
framework, which is where almost everything in .NET comes from. It
basically comes down to whether you want to use curly braces or not.
The difference in the two is that small. I haven't looked too much
into C++, and I haven't seen much documentation on C++.NET at all.

In my opinion, C++ is useful for three kinds of developers:
1) Ones who find that speed isn't satisfactory in other languages
2) People who need low-level access to the system
3) Gluttons for punishment. :)

Chris


--- In twsapi@y..., "dtkamp" <dkamp@e...> wrote:

In recent thread on IB board, Kent commented that C++ was the way

to

go for creating serious apps. Is this really true any more if

you're

starting from .NET, where all of its languages are based on same
underlying runtime lib.s and architecture. For purposes of building

a

killer trading app, does it really matter much whether you're using
VB.NET or C#.NET or C++? And can't you just mix in legacy C and C++
lib.s as necessary?

Thanks for your e-mail. I will be out of town for 3 days and will respond after returning. Please direct future e-mails to??????????? kybryan@...?????????????? thanks

marinindextrader wrote:

Another comment from the VB Explorer Board:

Re: C++ vs. C#.NET vs. VB.NET
----------------------------------------------------------------------
----------
Kent's full of crap! :)

I have yet to see a business app that can't be written in VB instead
of C++ and have the user know the difference. Most of the problems
with business apps involve query times against backend databases,
which usually isn't the front end's fault.

Just so you don't think I'm biased, I've done VB and C++ stuff for
almost 10 years. VB beats C++ hands down as the business dev language
of choice.



--- In twsapi@y..., "dtkamp" <dkamp@e...> wrote:

In recent thread on IB board, Kent commented that C++ was the way

to

go for creating serious apps. Is this really true any more if

you're

starting from .NET, where all of its languages are based on same
underlying runtime lib.s and architecture. For purposes of building

a

killer trading app, does it really matter much whether you're using
VB.NET or C#.NET or C++? And can't you just mix in legacy C and C++
lib.s as necessary?

Thanks Scott for getting responses.

Seems like way to go in a cooperative project is to get participants
to use .NET framework, where language choice is least likely to cause
problems, and C/C++ can be used to generate/process any necessary
unmanaged (= not under control of .NET runtime) or legacy code. I
think that will get us the most bang for the buck with repect to our
individual uses of the code (but perhaps not the most bang if someone
wanted to sell and distribute a cross-platform app - not a high
priority for me - but makes sense for IB's use of Java in TWS).

So what do the other 180 people up here think of what I just wrote
with respect to development platform selection for such a cooperative
effort? (We can do a separte survey to figure out who's actually
interested in such a project. Am just fishing for .NET objections to
start with.) Thanks!

Re: C++ vs. C#.NET vs. VB.NET
--------------------------------------------------------------------

--

----------
Kent's full of crap! :)

I have yet to see a business app that can't be written in VB

instead

of C++ and have the user know the difference. Most of the problems
with business apps involve query times against backend databases,
which usually isn't the front end's fault.

Just so you don't think I'm biased, I've done VB and C++ stuff for
almost 10 years. VB beats C++ hands down as the business dev

language

of choice.

The problem with .NET is that you have to distribute a 20MB .NET
runtime to everyone. Even Windows XP does not have the .NET runtime
included by default.

Monday, June 10, 2002, 2:50:02 PM, you wrote:

d> Thanks Scott for getting responses.

d> Seems like way to go in a cooperative project is to get participants
d> to use .NET framework, where language choice is least likely to cause
d> problems, and C/C++ can be used to generate/process any necessary
d> unmanaged (= not under control of .NET runtime) or legacy code. I
d> think that will get us the most bang for the buck with repect to our
d> individual uses of the code (but perhaps not the most bang if someone
d> wanted to sell and distribute a cross-platform app - not a high
d> priority for me - but makes sense for IB's use of Java in TWS).

d> So what do the other 180 people up here think of what I just wrote
d> with respect to development platform selection for such a cooperative
d> effort? (We can do a separte survey to figure out who's actually
d> interested in such a project. Am just fishing for .NET objections to
d> start with.) Thanks!

Re: C++ vs. C#.NET vs. VB.NET
--------------------------------------------------------------------

d> --

----------
??? Kent's full of crap! :)

I have yet to see a business app that can't be written in VB

d> instead

of C++ and have the user know the difference. Most of the problems
with business apps involve query times against backend databases,
which usually isn't the front end's fault.

Just so you don't think I'm biased, I've done VB and C++ stuff for
almost 10 years. VB beats C++ hands down as the business dev

d> language

of choice.

d> Yahoo! Groups SponsorADVERTISEMENT

d> To unsubscribe from this group, send an email to:
d> twsapi-unsubscribe@...



d> Your use of Yahoo! Groups is subject to theYahoo! Terms of Service.

Hi Dion. This could be fun. Let's get all the .NET negatives out in
the open. I guess some counterpoints to the runtime objection would
be,

- 20MB is (or is fast becoming) small potatoes, MS has made it easy
to get the .NET runtime update, and future versions of Windows will
include it

- if one's primary interest is in gaining access to a group's work
for one's own purposes (for trading), then one wouldn't be too
concerned that others might have a harder time getting it (and all
the developers get the runtime as part of their installation)

It is possible to waste years debating Java vs. Microsoft or VB vs.
C++ etc., and the debate doesn't get you anywhere. I view this forum
as a way to share code, learn from what others are doing and perhaps
use as a lobbying jumping off point to get the IB developers to
prioritize important API improvements / bug fixes. I don't want to
cooperatively work on a project to develop a killer app, and I'd
rather the list didn't get hung up in that either. I have my own app
and have no interest in being a cog in the next open source project.
I develop my trading software to fill a need in my trading. I'm sure
there are many others who do the same thing. Programming isn't my
primary focus, trading is. To the extent that I can improve my
trading by programming the IB TWS API, I will do it. To the extent
that I can offer code and help to others traveling the same path as
me, I will do it. To the extent that I can get help from professional
programmers to meet this end I am appreciative.

--- In twsapi@y..., "dtkamp" <dkamp@e...> wrote:

Thanks Scott for getting responses.

Seems like way to go in a cooperative project is to get

participants

to use .NET framework, where language choice is least likely to

cause

problems, and C/C++ can be used to generate/process any necessary
unmanaged (= not under control of .NET runtime) or legacy code. I
think that will get us the most bang for the buck with repect to

our

individual uses of the code (but perhaps not the most bang if

someone

wanted to sell and distribute a cross-platform app - not a high
priority for me - but makes sense for IB's use of Java in TWS).

So what do the other 180 people up here think of what I just wrote
with respect to development platform selection for such a

cooperative

effort? (We can do a separte survey to figure out who's actually
interested in such a project. Am just fishing for .NET objections

to

start with.) Thanks!

Hi tripack. I'm sure your description fits most of IB's traders who
use the API. A few of us are just trying to think a bit outside the
lone wolf box. With respect to your reference to "open source",
Kent's proposal suggested that anyone involved in development or
funding would have ownership or use rights - no open source would be
involved as the term is commonly understood.

--- In twsapi@y..., tripack44 <no_reply@y...> wrote:

It is possible to waste years debating Java vs. Microsoft or VB vs.
C++ etc., and the debate doesn't get you anywhere. I view this

forum

as a way to share code, learn from what others are doing and

perhaps

use as a lobbying jumping off point to get the IB developers to
prioritize important API improvements / bug fixes. I don't want to
cooperatively work on a project to develop a killer app, and I'd
rather the list didn't get hung up in that either. I have my own

app

and have no interest in being a cog in the next open source

project.

I develop my trading software to fill a need in my trading. I'm

sure

there are many others who do the same thing. Programming isn't my
primary focus, trading is. To the extent that I can improve my
trading by programming the IB TWS API, I will do it. To the extent
that I can offer code and help to others traveling the same path as
me, I will do it. To the extent that I can get help from

professional

programmers to meet this end I am appreciative.

--- In twsapi@y..., "dtkamp" <dkamp@e...> wrote:

Hi Dion. This could be fun. Let's get all the .NET negatives out in
the open. I guess some counterpoints to the runtime objection would
be,

.net does not support DDE, in some applications(perhaps all). I
have limited research is there a expert that could enlighten us.